Skip to main content

Dialysis modality and readmission following hospital discharge: a population-based cohort study

Perl J, McArthur E, Bell C, Garg AX, Bargman JM, Chan CT, Harel S, Li L, Jain AK, Nash DM, Harel Z. Am J Kidney Dis. 2017; 70(1):11-20. Epub 2017 Jan 6.

Background — Readmissions following hospital discharge among maintenance dialysis patients are common, potentially modifiable, and costly. Compared with patients receiving in-center hemodialysis (HD), patients receiving peritoneal dialysis (PD) have fewer routine dialysis clinic encounters and as a result may be more susceptible to a hospital readmission following discharge.

Study Design — Population-based retrospective-cohort observational study.

Settings and Participants — Patients treated with maintenance dialysis who were discharged following an acute-care hospitalization during January 1, 2003, to December 31, 2013, across 164 acute-care hospitals in Ontario, Canada. For those with multiple hospitalizations, we randomly selected a single hospitalization as the index hospitalization.

Predictor — Dialysis modality PD or in-center HD. Propensity scores were used to match each patient on PD therapy to 2 patients on in-center HD therapy to ensure that baseline indicators of health were similar between the 2 groups.

Outcome — All-cause 30-day readmission following the index hospital discharge.

Results — 28,026 dialysis patients were included in the study. 4,013 PD patients were matched to 8,026 in-center HD patients. Among the matched cohort, 30-day readmission rates were 7.1 (95% CI, 6.6-7.6) per 1,000 person-days for patients on PD therapy and 6.0 (95% CI, 5.7-6.3) per 1,000 person-days for patients on in-center HD therapy. The risk for a 30-day readmission among patients on PD therapy was higher compared with those on in-center HD therapy (adjusted HR, 1.19; 95% CI, 1.08-1.31). The primary results were consistent across several key prespecified subgroups.

Limitations — Lack of information for the frequency of nephrology physician encounters following discharge from the hospital in both the PD and in-center HD cohorts. Limited validation of International Classification of Diseases, Tenth Revision codes.

Conclusions — The risk for 30-day readmission is higher for patients on home-based PD compared to in-center HD therapy. Interventions to improve transitions in care between the inpatient and outpatient settings are needed, particularly for patients on PD therapy.

View full text

Keywords: Dialysis Risk assessment Hospitalization Kidney and urinary tract disorders